IMPLEMENTATION AND RESEARCH OF NEURAL NETWORKS OF DIRECT PROPAGATION ON FPGA

Authors

  • Volodymyr Shymkovych National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine
  • Yurii Berdnyk National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine

Abstract

This paper presents the results of FPGA implementation of direct propagation neuronal networks for the previously developed method of realization of nonlinear activation functions and artificial neurons implementation algorithms. The obtained results of the research, which are presented in the table, are FPGA source, error and speed.  

Key words: neural network, activation function, FPGA, VHDL.

Fig.: 1. Tabl.:1. Bibl.: 16.

Published

2023-11-08

Issue

2018

Section

IoT, Real Time Systems